DEEPBEAS3D: Deep Learning and B-Spline Explicit Active Surfaces

TL;DR

This paper introduces DEEPBEAS3D, a 3D deep learning framework using B-spline explicit active surfaces for improved, controllable, and efficient segmentation of anatomical structures in ultrasound images, addressing robustness and usability issues.

Contribution

It presents a novel 3D extension of an interactive segmentation method that combines CNNs with B-spline surfaces for better control and anatomical plausibility.

Findings

Reduced user workload by 30% compared to clinical tool

Achieved 70% less user time (170 seconds) for segmentation

Provided explicit control of the surface contour

Abstract

Deep learning-based automatic segmentation methods have become state-of-the-art. However, they are often not robust enough for direct clinical application, as domain shifts between training and testing data affect their performance. Failure in automatic segmentation can cause sub-optimal results that require correction. To address these problems, we propose a novel 3D extension of an interactive segmentation framework that represents a segmentation from a convolutional neural network (CNN) as a B-spline explicit active surface (BEAS). BEAS ensures segmentations are smooth in 3D space, increasing anatomical plausibility, while allowing the user to precisely edit the 3D surface. We apply this framework to the task of 3D segmentation of the anal sphincter complex (AS) from transperineal ultrasound (TPUS) images, and compare it to the clinical tool used in the pelvic floor disorder clinic (4D View VOCAL, GE Healthcare; Zipf, Austria). Experimental results show that: 1) the proposed framework gives the user explicit control of the surface contour; 2) the perceived workload calculated via the NASA-TLX index was reduced by 30% compared to VOCAL; and 3) it required 7 0% (170 seconds) less user time than VOCAL (p< 0.00001)